Re: [libvirt] [Libvirt-ci] Build failed in Jenkins: virt-manager-check » libvirt-fedora-rawhide #1056
On 9/3/19 9:31 AM, ci@centos.org wrote:
See <https://ci.centos.org/job/virt-manager-check/systems=libvirt-fedora-rawhide/1056/display/redirect>
------------------------------------------ Started by upstream project "virt-manager-check" build number 1056 originally caused by: Started by upstream project "virt-manager-build" build number 1060 originally caused by: Started by upstream project "libvirt-python-build" build number 685 originally caused by: Started by upstream project "libvirt-build" build number 799 originally caused by: Started by an SCM change Running as SYSTEM [EnvInject] - Loading node environment variables. Building remotely on libvirt-fedora-rawhide (libvirt) in workspace <https://ci.centos.org/job/virt-manager-check/systems=libvirt-fedora-rawhide/ws/> [virt-manager] $ /bin/sh -xe /tmp/jenkins12079261823599550029.sh + /usr/bin/python3 ./setup.py test running test Traceback (most recent call last): File "/home/jenkins/build/libvirt/lib64/python3.7/site-packages/libvirt.py", line 24, in <module> import cygvirtmod as libvirtmod ModuleNotFoundError: No module named 'cygvirtmod'
During handling of the above exception, another exception occurred:
Traceback (most recent call last): File "./setup.py", line 661, in <module> distutils.core.setup( File "/usr/lib64/python3.8/distutils/core.py", line 148, in setup dist.run_commands() File "/usr/lib64/python3.8/distutils/dist.py", line 966, in run_commands self.run_command(cmd) File "/usr/lib64/python3.8/distutils/dist.py", line 985, in run_command cmd_obj.run() File "./setup.py", line 528, in run TestBaseCommand.run(self) File "./setup.py", line 448, in run import tests as testsmodule File "<https://ci.centos.org/job/virt-manager-check/systems=libvirt-fedora-rawhide/ws/tests/__init__.py",> line 16, in <module> from virtinst import buildconfig File "<https://ci.centos.org/job/virt-manager-check/systems=libvirt-fedora-rawhide/ws/virtinst/__init__.py",> line 42, in <module> _set_libvirt_error_handler() File "<https://ci.centos.org/job/virt-manager-check/systems=libvirt-fedora-rawhide/ws/virtinst/__init__.py",> line 32, in _set_libvirt_error_handler import libvirt File "/home/jenkins/build/libvirt/lib64/python3.7/site-packages/libvirt.py", line 28, in <module> raise lib_e File "/home/jenkins/build/libvirt/lib64/python3.7/site-packages/libvirt.py", line 20, in <module> import libvirtmod ModuleNotFoundError: No module named 'libvirtmod' Build step 'Execute shell' marked build as failure
Anyone know what's going on here? I don't think this is virt-manager's fault. The console output from the latest libvirt-python build: https://ci.centos.org/job/libvirt-python-build/685/systems=libvirt-fedora-ra... Shows libvirt.py and libvirtmod.py being put into /home/jenkins/build/libvirt/lib64/python3.8/site-packages , but the above traceback indicates there's a libvirt.py in /home/jenkins/build/libvirt/lib64/python3.7/site-packages (3.8 vs 3.7). Maybe there's stale libvirt.py in the rawhide image? - Cole - Cole
On Tue, Sep 03, 2019 at 10:23:08AM -0400, Cole Robinson wrote:
Traceback (most recent call last): File "/home/jenkins/build/libvirt/lib64/python3.7/site-packages/libvirt.py", line 24, in <module>
[snip]
Anyone know what's going on here? I don't think this is virt-manager's fault.
The console output from the latest libvirt-python build: https://ci.centos.org/job/libvirt-python-build/685/systems=libvirt-fedora-ra...
Shows libvirt.py and libvirtmod.py being put into /home/jenkins/build/libvirt/lib64/python3.8/site-packages , but the above traceback indicates there's a libvirt.py in /home/jenkins/build/libvirt/lib64/python3.7/site-packages (3.8 vs 3.7). Maybe there's stale libvirt.py in the rawhide image?
The PYTHONPATH env var (and all other env vars) are set at the OS level which means they are fixed at the time the jenkins slave starts. We periodically upgrade RPMs via dnf, and this time python got upgraded to 3.8. The slave's PYTHONPATH still says 3.7 though. I'll reboot the VM to fix it shortly. Really we should be setting all build related env vars in the job script, so that they are guaranteed accurate when each job runs. Regards, Daniel -- |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| |: https://libvirt.org -o- https://fstop138.berrange.com :| |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|
On Tue, 2019-09-03 at 10:23 -0400, Cole Robinson wrote:
On 9/3/19 9:31 AM, ci@centos.org wrote:
See <https://ci.centos.org/job/virt-manager-check/systems=libvirt-fedora-rawhide/1056/display/redirect>
------------------------------------------ Started by upstream project "virt-manager-check" build number 1056 originally caused by: Started by upstream project "virt-manager-build" build number 1060 originally caused by: Started by upstream project "libvirt-python-build" build number 685 originally caused by: Started by upstream project "libvirt-build" build number 799 originally caused by: Started by an SCM change Running as SYSTEM [EnvInject] - Loading node environment variables. Building remotely on libvirt-fedora-rawhide (libvirt) in workspace <https://ci.centos.org/job/virt-manager-check/systems=libvirt-fedora-rawhide/ws/> [virt-manager] $ /bin/sh -xe /tmp/jenkins12079261823599550029.sh + /usr/bin/python3 ./setup.py test running test Traceback (most recent call last): File "/home/jenkins/build/libvirt/lib64/python3.7/site-packages/libvirt.py", line 24, in <module> import cygvirtmod as libvirtmod ModuleNotFoundError: No module named 'cygvirtmod'
During handling of the above exception, another exception occurred:
Traceback (most recent call last): File "./setup.py", line 661, in <module> distutils.core.setup( File "/usr/lib64/python3.8/distutils/core.py", line 148, in setup dist.run_commands() File "/usr/lib64/python3.8/distutils/dist.py", line 966, in run_commands self.run_command(cmd) File "/usr/lib64/python3.8/distutils/dist.py", line 985, in run_command cmd_obj.run() File "./setup.py", line 528, in run TestBaseCommand.run(self) File "./setup.py", line 448, in run import tests as testsmodule File "<https://ci.centos.org/job/virt-manager-check/systems=libvirt-fedora-rawhide/ws/tests/__init__.py",> line 16, in <module> from virtinst import buildconfig File "<https://ci.centos.org/job/virt-manager-check/systems=libvirt-fedora-rawhide/ws/virtinst/__init__.py",> line 42, in <module> _set_libvirt_error_handler() File "<https://ci.centos.org/job/virt-manager-check/systems=libvirt-fedora-rawhide/ws/virtinst/__init__.py",> line 32, in _set_libvirt_error_handler import libvirt File "/home/jenkins/build/libvirt/lib64/python3.7/site-packages/libvirt.py", line 28, in <module> raise lib_e File "/home/jenkins/build/libvirt/lib64/python3.7/site-packages/libvirt.py", line 20, in <module> import libvirtmod ModuleNotFoundError: No module named 'libvirtmod' Build step 'Execute shell' marked build as failure
Anyone know what's going on here? I don't think this is virt-manager's fault.
The console output from the latest libvirt-python build: https://ci.centos.org/job/libvirt-python-build/685/systems=libvirt-fedora-ra...
Shows libvirt.py and libvirtmod.py being put into /home/jenkins/build/libvirt/lib64/python3.8/site-packages , but the above traceback indicates there's a libvirt.py in /home/jenkins/build/libvirt/lib64/python3.7/site-packages (3.8 vs 3.7). Maybe there's stale libvirt.py in the rawhide image?
Pavel looked into it earlier and he thinks he might be a result of the Jenkins agent having an old $PYTHONPATH in its environment, so a reboot should fix it. I plan to upgrade and reboot the host tomorrow, let's see whether it persists after that. -- Andrea Bolognani / Red Hat / Virtualization
participants (3)
-
Andrea Bolognani -
Cole Robinson -
Daniel P. Berrangé